Just a heads up: On March 24, 2025, starting at 4:30pm CDT / 19:30 UTC, the site will be undergoing scheduled maintenance for a few hours. During this time, the site might be unavailable for a short while. Thanks for your patience.
×I have a plan that consists of two jobs, the first running on Linux, the second on Windows. The second job needs the artifact created by the first job.
Right now my Windows job has 3 tasks: remove all files ("del *.tar.gz"), download the artifact, then a script do its real work. I would like to combine those into 1 script task (because I have dozens of plans like these, and I'm tired of maintaining them all separately in Bamboo). That means I need to somehow grab the artifact in my script.
Is there a way to do this? Can a URL for the artifact be passed to the script or something?
Failing that, perhaps I can keep the artifact-download task in Bamboo, and then my script can delete all files except that artifact, but that requires knowing what that artifact file is called, something I'm not seeing a variable for.
Another solution would be to click the "Clean working directory after each build" box for this stage, but I don't like that because I often have to go look at the directory after a failed build and figure out what went wrong. I wish there were a "Clean working directory before each build" option.
I'm running Bamboo version 5.12.2 build 51212, if there's something in a more recent version of Bamboo that would help, we're open to upgrading.
Jobs are run concurrently so that's not the way to do it.
You must move second job to second stage or move it to second plan. I
If you use stage you can use job artifact subscription.
If you use second plan you can use download artifacts task.
Ok I think I understand your question better now.
In general you can't get artifact url in script task.
There are two good reasons for that:
If you could assume some artifact name as a standard then this one would be possible: https://jira.atlassian.com/browse/BAM-2357 (last comment)
Artifact subscription is called "dependency" in documentation. Take a look at https://confluence.atlassian.com/display/BAMBOO/Sharing+artifacts in section Sharing artifacts between jobs.
What you could do is to create this dependency in second job and put it into some specific directory. Then in script you could delete everything except that directory.
Trouble is I'm not sure if it makes things much simpler. You can have one task, but you need to create this dependency in job...
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Sorry, I get confused with the terminology - these two jobs are indeed in separate stages. They are each the only job in their stage.
Can you say more about "artifact subscription"? I couldn't find that terminology in the docs. Currently I'm using the download artifacts task.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.